Artifacts, Archaeology, and Cabeza De Vaca in Southern Texas and Northeastern Mexico
by Thomas R. Hester, The University of Texas at Austin
Of all the various routes published for Cabeza de Vaca over the past century, none make any sense, in terms of the archaeological record, except those of Krieger (1961) and Campbell and Campbell (1981). This route (see also Chipman 1987: 143,145), hereafter referred to as the "southern route," takes Alvar Núñez Cabeza de Vaca and his group from the upper and central Texas coast, across southern Texas, and into northeastern Mexico. Since the archaeological record of this region has existed for only a few decades, earlier scholars could not have taken full advantage of the Cabeza de Vaca chronicles in considering the various alternatives for the route. Even today, we can push the "southern route" a bit further south than Krieger, Campbell and Campbell because of what Cabeza de Vaca did not record and what we have learned archaeologically about the region in the past 30 years. Most importantly, in this planning document and in the Witte Museum/Texas State University planning meetings, we can be fairly confident of this "southern route" and use it to predict the kinds of cultures and cultural materials that Cabeza de Vaca encountered. Then, appropriate artifacts and archaeological cultures can be selected from extant collections for interpretative exhibits.
Cabeza de Vaca and Coastal Native Groups
Based on archaeological studies by Robert Ricklis on Galveston Island, new insights can be added to the initial phase of Cabeza de Vaca's capture and subsequent movements among coastal peoples. There is little doubt that the Narváez expedition, of which Cabeza de Vaca was a member, was shipwrecked on Galveston Island or a nearby area of the upper Texas coast. The descriptions of the peoples they encountered closely match the archaeological remains found by Ricklis, though they are perhaps broadly representative of upper coastal cultural adaptations in the early 16th century. Cabeza de Vaca records the use of the bow and arrow, fish weirs or traps, and a heavy reliance both on fish and roots during fall and winter on the island. Special mortuary treatment was reserved for children and for "medicine men," who were cremated.
Though the current political situation prevents us from using any burial materials in a museum exhibit, Ricklis found that most grave goods were lavished on adult males and children. One adult burial contained what Ricklis (1994:476) describes as a "rat‑tooth bloodletting instrument," which he infers to be a "shaman's possession," albeit not with a cremation. If one looks at the Cabeza de Vaca chronicle, the reference to the use of such an item in bloodletting involved a child (probably in northern Mexico or in the Southwestern United States), and the bloodletting, with the use of "sharp mouse teeth" (Covey 1961: 113), was apparently not done by a shaman or medicine man. I will return to this matter at the end of this paper.
Ricklis' excavations also produced examples of dwellings (post-mold patterns and artifact concentrations) that would be more in line with a long‑term occupation by a sizable group of up to several hundred people, perhaps in the fall‑winter situation described by Cabeza de Vaca, although he did not specifically describe their houses. Whether these peoples were Karankawa, Akokisa, or some other group, they are more complex in term of social organization and belief systems than Cabeza de Vaca recorded. Clearly, some of the domestic artifacts dating to the early 16th century from Ricklis' excavations, now housed at the Texas Archeological Research Laboratory (TARL) at The University of Texas at Austin, would be especially appropriate for exhibit. These include arrow points and other stone tools, bone and shell artifacts, and pottery (Covey 1961:55 notes that an "earthen pot" was one of the objects stolen from a native hut on the island; a recent translation of Cabeza de Vaca by Maria Wade of the UT Austin Department of Anthropology indicates that this "pot" was an olla). In addition, reconstructions of house floors (post-mold patterns) and food processing areas (hearths) could be done. As I noted above, use of burial goods or even photographs of burials/burial goods would likely result in complaints from Native American groups.
It is during Cabeza de Vaca's presence among the coastal groups that we learn the most about material culture, especially the kinds that survive archaeologically. This is the role of trader that Cabeza de Vaca began in 1530 when he lived among the "people of Charruco" in the coastal zone south of Galveston Island. Much has been written about Cabeza de Vaca's role as a trader (e.g., Corgan 1969; Wade 1996), but here I will focus specifically on the material goods that he exported and imported, and what that tells us archaeologically.
Wade (personal communication, 1996), in her translation of Cabeza de Vaca, indicates that the items he took with him from the coast were pieces of "sea snails" (conch or other gastropods) and their "hearts" (here, he must be referring to the columella of the conch, widely prized as a raw material), and shells to cut a fruit‑like bean with which they cure and make bailes and fiestas (this "bean" would not have survived archaeologically, though a discussion of the identify of this vegetal item would be important).
With these goods, Cabeza de Vaca was able to travel into the interior, apparently crossing territories without conflict. I think this is the result of a trading system of great time depth in southern Texas, in which he just happened to get involved. There is ample archaeological evidence, on which I have written a number of times, of pre-Hispanic trade in southern Texas (cf. Hester 1971a,b, 1980, 1995).
Once in the interior, he traded his coastal goods for the following items, again based on Wade's translation: "skins," "red dirt with which they grease and paint their faces and hair" (hematite or red ocher), chert (pedernales) points of arrows, hard cane to make arrow shafts (flechas), glue, and "some tassels which they make of skin of deer and dye red."
Cabeza de Vaca's trading endeavors closely match the archaeological record of the central Texas coast and southern Texas, perhaps even into central Texas and the lower Pecos. Covey's (1961) suggestions of his incursions going into East Texas and Oklahoma can be discounted. The Texas coast was without chippable stone (chert or "flint"), and since archaic times, there is ample evidence of south and central Texas cherts being obtained by coastal peoples to make spear and arrow points. This may have even accelerated in Late Prehistoric times, roughly the era of Cabeza de Vaca, when bison hunting became important on the coastal prairies,and the need for more chert for arrow points, end scrapers, knives, and perforators increased (Hester and Parker 1970). Chert of pretty good quality is available on the lower Nueces and the lower Guadalupe, 15-50 miles upstream from the coast (Chandler 1984). Red ochre or hematite appears in coastal burials (e.g., Hester 1969), as well as in those on the coastal/interior eco-tone (not sure?), such as the Archaic cemetery at Loma Sandia (Taylor and Highley 1995). Skins, arrow shafts, and tassels have left no archaeological trace. I am not sure what "glue" the coastal peoples desired from the interior, perhaps sap from mesquite or similar trees. This sap makes a good mastic. However, they had available, and had used since at least 600‑800 B.C., asphaltum that washed up on the beaches and could be melted to use as an adhesive.
In the interior, marine shell is present at many sites (cf. Hester 1971 a,b, 1980, Black 1986; Highley 1986). Indeed, ornaments of marine shell imported from the Texas coast are found as early as 5800 years ago at the Bering Sink mortuary site in Kerr County (Bement 1994). Marine shell usually occurs as scattered specimens, especially conch shell fragments, conch columella (often made into either tools or disc‑shaped beads), and fragments of ribbed cockle shell such as Dinocardium robustum. Here, again, it is my distinct impression that the very late Late Prehistoric peoples had more of this material, as at the Hinojosa site near Alice (Black 1986) and site 41LK201 on the Frio River, west of Three Rivers (Highley 19860). An alternative to trade would be interaction between coastal and interior groups. Ricklis (1996) has suggested contemporary cohabitation of a coastal margin site, 41RF21, and perhaps during such events, trade would have taken place. But given the time depth of such trade in the interior of south, central, and lower Pecos Texas, I suspect that some other mechanism led to an intensification of exchange in the Late Prehistoric period. Certainly, once shell artifacts had been obtained by interior groups, they could have been "redistributed" through trade during the annual prickly pear harvests.
Cabeza de Vaca in the Interior
Cabeza de Vaca's activities on the margin of the coastal prairies and his subsequent journey through the interior to the Rio Grande raise several archaeological issues. For example, references to subsistence activities are of interest. While with the Mariames on the lower Guadalupe River (we assume), Cabeza de Vaca and his colleagues were involved in the processing of pecans. This may have involved ground stone manos and metates, or even the use of stone pestles. Alternatively, they may have used wooden mortars and pestles of the sort reported ethno-historically in Nuevo Leon in the 17th century by Alonso de Leon (Brown 1988; see also Beals 1932), and archaeologically, from the lower Pecos area by Collins and Hester (1968) and Prewitt (1981). In addition, the harvesting and processing of tuna, the fruit of prickly pears, might have involved similar gear or an earthen pit.
It is of great interest to me that when, in 1532, Cabeza de Vaca and colleagues are involved in the summer harvesting of prickly pear tuna, there is no mention of pottery, which would have been ideal for the storage of beverages made from tuna juices (cf Krieger 1956:53). Indeed, he mentions a "lack of vessels" for the juice from the tunas. Either this is specific only to the group with whom Cabeza de Vaca's was associated, or his route across southern Texas was a bit farther south than the "southern route" currently suggests. I can say this with confidence in that the peoples of the southern Texas interior had begun making bone-tempered pottery, including olla forms, as early as A.D. 1200-1300 (cf. Hester 1980:126; Highley 1986) and continued making this ware through the Late Prehistoric Period during the time they were residing in the 18th-century Spanish missions. Pottery was being made as far south as Alice, and clearly north of a line running from Dimmit County down to Alice and eastward to Baffin Bay. Thus, the map of his journey might be more accurately drawn as running south of Alice (per Campbell and Campbell 1981 : Figure l), – or his route took a decided southern turn per Wells and Davenport (cf. Chipman 1987:131). I doubt the latter is the case, as the Wells and Davenport route would have put the castaways in the South Texas Sand Sheet area, a patch of south Texas that Cabeza de Vaca would have surely remembered. Indeed, if Cabeza de Vaca was among the Avavares, whose territory included the legendary great prickly pear fields, he would then have been in Webb, Duval, and southern Jim Wells County, south of the pottery‑making peoples.
While among the Avavares, Cabeza de Vaca was told of the myth of Mala Cosa, the Bad Thing. Among his deeds was the use of a "flint knife two palms long and a hand wide" (Covey 1961:90) to inflict wounds on hapless individuals. There is no raw material in south Texas that would have allowed the Avavares to even envision a stone tool of that size (the Uvalde and Rio Grande gravels used for points and everyday tools are relatively small cobbles). A description of such a massive biface (knife) may reflect what we see archaeologically in the region; that is, the presence of large, thin bifaces that have been traded into the region from the Edwards Plateau (e.g. Taylor and Highley 1995, Hester and Barber 1990). Such specimens are in both the Witte and TARL collections.
Cabeza de Vaca's projected entry into Mexico, near what is today the Falcon Reservoir district, is probably accurate. We are still learning about the Late Prehistoric peoples in that region, and the receipt of radiocarbon dates on several burials (with associated grave goods) may shed new light on the people of that region in the 16th century (cf. Hester 1995).
It should also be noted that this area, especially Starr, Zapata, and Jim Hogg Counties, has a high density of prickly pear, but little in the way of creeks or other surface water (Hester 1981). This area and the area to the east could have been the territories of the Avavares (at least seasonally), and certainly the Arbadaos (Campbell and Campbell 1981). The dire descriptions of their subsistence, during Cabeza de Vaca's eight months in the monte, would certainly fit with the modeled settlement pattern and low-density resource that I have put forth for the region (Hester 1981; why Covey (1961 :92) would also equate this locale with the Texas "hill country" is indeed a mystery). Cabeza de Vaca's stay among the Cuchendados (Campbell and Campbell 1981:393) on the Rio Grande demonstrates his adaptation to life in the region (Wade 1996) and his skills at making items of material culture, such as mats, which were greatly prized as house coverings in what appear to have been at least a couple of sizable villages.
Cabeza de Vaca offers other insights into subsistence that have archaeological implications. He notes bison in the area of the Mariame, which Covey, for some inexplicable reason, places in the area of Austin, and clearly there were bison at this period on the coastal prairies (cf. Hester and Parker 1970, Ricklis 1996). He also points to the fact that the Mariame and related groups ate practically everything. This is borne out by the long lists of faunal remains at Late Prehistoric sites in the region (e.g., Black 1986, Highley 1986; Hester and Hill 1975). It is intriguing that he notes the pulverization of bone, to reduce these materials to bone grease to be mixed with other foods. This is a well known practice among southern California Indians (Bean 1972:66). However, faunal remains are so extensive and so well preserved at most Late Prehistoric sites in south Texas that this technology must have been used either with certain species or at certain times of the year or in cases of famine. Krieger (1956:53) reports that snails (land, or prairie snails; Rabdotus sp.) were also eaten, though, I cannot locate this reference in the published translations. Campbell (1983) also reports the eating of snails by the Mariame while they were in the tuna fields.
The archaeological evidence is overwhelming in terms of snail consumption. Along the central and southern coast, on the coastal prairies, and into the interior, Late Prehistoric sites have abundant land snails in the middens. Some archaeologists attribute their presence to the scavenging activities of land snails, but this is a facile argument that ignores the archaeological data (e.g., their association with hearths and food processing areas; cf Scott 1982; Black 1986, Highley 1986). If the snails were incorporated into the middens as casual scavengers, then they have clearly changed their feeding habits in South Texas since late prehistory (Hester 1995). Snail shells are so common at sites in Starr and Zapata Counties, the territory of the Arbadaos (and Cuchendados?) that archaeological sites can be easily recognized during survey while riding in a truck by the glistening white sheen of the mass of snails eroding out of the sites (Nunley and Hester 1975).
Cabeza de Vaca refers to the digging of roots. This would have been done with digging sticks, of the kind that survive today only in dry caves (e.g., examples from the Lower Pecos caves in the Witte Museum collections). He also describes the use of earth ovens for cooking roots. Archaeologists working in the region have not conclusively identified such pits, though there are a variety of hearths, including charcoal and rock-filled features, that might have served in this fashion (as at Choke Canyon, Scott 1982, Hester et al. 1975 report deep rock filled pits from McMullen County that were clearly earth ovens). While Cabeza de Vaca disparages the Indians' use of fire to burn off the prairies, this was a widespread practice in Texas and the Great Plains. Archaeological evidence of such activities is, of course, hard to come by, though McGraw (1983:91) recorded thin lines of charcoal in excavation profiles near Laredo that might reflect such activities.
I have always been bothered by Cabeza de Vaca's descriptions of hunger, famine, and low quantities of food just about everywhere he went (these descriptions have been embellished; cf. Krieger 1956). This simply does not fit with the archaeological evidence, in terms of faunal remains (animals, fish, marine or freshwater shell, snails, etc.) that occur in sites in the area where he must have traveled. Perhaps his eight months in the monte in the south Texas interior reflected seasonal variations in food availability, as recorded by Alonso del Leon (Brown 1988). Or perhaps, he was there in a time of drought. Having been in the area suffering from a significant drought, I can imagine the shortage of resources he encountered. Or, perhaps, his descriptions reflect his bias as a European who had to subsist on unfamiliar and distasteful foods.
The settlement pattern described for the Mariame and associated groups, to be near wood and water, fits what is seen for the placement of prehistoric occupation sites in the region (Cabeza de Vaca later contradicts himself by noting that the Indians "have no familiar places for getting water") (Covey 1961: 83). Perhaps here he referred to groups deeper into south Texas, where creeks and water sources were indeed fewer (the low‑density zones described by Hester 1981). The brevity with which the sites were occupied likely suggests repeated occupations over the years, again fitting with the "settlement zones" seen in the areas of higher density resources in the region (Hester 1981). Some sites were apparently occupied long enough for houses, albeit flimsy ones, to be set up. Archaeological evidence for such houses is limited to a few postmolds or trash distribution patterns from a few sites (cf. Black 1986: 266).
To summarize at this point, museum material culture appropriate to Cabeza de Vaca during his "trader" phase and his journey thence to the Rio Grande would include arrow points (likely Perdiz, and perhaps Caracara at Falcon Lake), bone‑tempered pottery (including restored or partially restored vessels at TARL) and, from Choke Canyon, at the Center for Archaeologica Research at The University of Texas at San Antonio, "hide scrapers" (end scrapers) and beveled knives (bison processing equipment), large trade bifaces, manos and metates, mats (as among the Cuchendado, use Lower Pecos examples), bows and arrows, digging sticks, etc. (also derived from Lower Pecos collections but surely identical to what was in the south Texas interior). Photographs of the wooden mortars and pestles reported by Collins (1968) and Prewitt (1981) would be appropriate, given Cabeza de Vaca's description of mesquite bean processing (the Collins and Hester specimen is from the Lower Pecos, and had prickly pear seeds in the mortar) on the Rio Grande. He clearly describes pestles, and mentions a "jar" which sounds much like the mortars that have been reported.
Cabeza de Vaca in Northeastern Mexico
When Cabeza de Vaca moves west from the Rio Grande, he entered territory that is still very poorly known archaeologically, especially in Late Prehistoric times. No major excavations or studies have dealt with this terminal portion of the culture sequence in the region, and those that have usually involved rockshelters. (cf Hester et al. 1994)(not part of Cabeza de Vaca's chronicle.) Cabeza de Vaca's reference to rabbit hunting sticks clearly refer to the "rabbit sticks" present in museum collections at the Witte and at TARL, from the lower Pecos and western Texas. Doubtless, they are the same as he recorded in northeastern Mexico.
However, there is a much discussed incident involving material culture that I would like to touch upon. This is the "copper" bell or rattle (described as "squat and wide;" M. Wade, personal communication, 1996) given to Dorante "at the base of the mountains" (Covey 1961:110; Ketchum 1988; Epstein 1991). Cabeza de Vaca relates that the gift-givers had obtained the artifact from the "north." This has led many to speculate that it is derived from Casas Grandes, the great trading and craft production center in Chihuahua studied by Di Peso (cf. Di Peso l974, who notes Casas Grandes trade networks reaching to La Junta de los Rios and other northern Mexico locales). It may have indeed come from there, but I suggest an alternate scenario.
If the Krieger map is correct (cf Chipman 1987:143), this event probably took place south of Monterrey. This is only 250 miles from the Pavon site, on the Rio Panuco, in the Huastecan culture area. At this site, Ekholm (1944:478‑479) reports and illustrates large copper bells (including one strikingly similar to a specimen from Casas Grandes, DiPeso 1974 510) and notes several others, including one in the form of a turtle in the Huasteca. The one given Dorante apparently had a "face" or image on it. Indeed, Huastecan outposts in the Sierra de Tamaulipas, occupied at the time of Cabeza de Vaca (the Mesoamerican Late Postclassic) were within 100 miles. The Huastecans were very active traders, as seen in their interaction with the hunters and gatherers of the Brownsville complex in the Rio Grande delta, where obsidian, pottery, and jade is documented (see Hester 1995 for a detailed review). Furthermore, despite all the ruminations in the various translations, could the "South Sea" and its riches be not the Pacific or the Bay of California… but the Gulf of Mexico with its Huastecan towns and traders?
Finally, I want to close with some observations on the material culture related to the famed "mouse‑teeth" scarification event (Covey 1961: 113). It is hard to say where this took place, though Krieger's route suggests it could have been in the Big Bend area. Of particular relevance here is a basket and contents found at site 41VV171 in the 1930s and in the collections at TARL. This has never been fully published, though it is sometimes called a "shaman's kit" or a "medicine kit." Among the contents are mescal beans (known hallucinogens) and a number of rodent (rabbit) mandibles. It is not hard to imagine these latter items being used for this type of scarification, especially in the scene described in the Cabeza de Vaca case. Though Cabeza de Vaca describes the event as one of parents punishing a crying infant, one could suspect that it is related to curing rituals. If so, such scarification appears not to have been solely the domain of a "shaman," (cf Ricklis 1994:416) but could be done by anyone. I do not want to stretch this too much, but simply point to this kit (or a photograph of it, again due to Native American concerns) as a possible exhibit item, given the mention of the scarification event and, of course, Cabeza de Vaca's role as a healer during much of his time in Texas and Mexico.